Substation risk early warning method and system based on edge AI analyzer

Abstract

The invention relates to the technical field of risk early warning, and particularly discloses a substation risk early warning method and system based on an edge AI analyzer, and the method comprises the steps: determining the position of the edge AI analyzer and the application range of the edge AI analyzer according to a determined gunlock position; the method comprises the following steps: receiving a video acquired by a gun camera regularly based on an edge AI analyzer, performing target positioning on the video, determining a target frame, identifying contents in the target frame based on a preset convolution feature table, and outputting a risk type; and when the output of identifying the content in the target frame based on the preset convolution feature table is null, identifying the content in the target frame based on an AI identification model in the edge AI analyzer, and outputting a risk type. The local identification module based on the convolution kernel table and the AI identification module based on the AI model are arranged in the edge AI analyzer, the transformer substation risk is identified in a gradient mode, the identification speed is high, and the identification range is large.

Description

Technical Field

[0001] The present invention relates to the field of risk warning technology, and specifically to a substation risk warning method and system based on an edge AI analyzer. Background Art

[0002] Visual AI Edge Analyzer System - The edge analyzer is an AI model computing device with visual AI model reasoning capabilities. It extracts frames from the video stream, inputs continuous images into the convolutional neural network for image recognition, and combines business logic programs for logical judgment, ultimately generating alarm events.

[0003] There are many application scenarios of visual AI edge analyzers, including mining, coal, oil, petrochemical and electricity. The application scenario of this application is the power field. The existing power risk warning process relies more on manual labor. As shown in the patent with announcement number: CN112734254B, risk warning relies more on manual labor, that is, determine the current target construction process to be carried out from the preset three-dimensional construction model, obtain the risk point information corresponding to the target construction process, generate risk warning information according to the risk point information, and send the risk warning information to the terminal device. Although an intelligent construction process determination scheme is introduced in this process, the level of intelligence is not high, and the work of risk monitoring personnel is extremely boring. In today's rapid development of AI technology, how to use AI technology to intelligently optimize inspection tasks is the technical problem that the present invention wants to solve. Summary of the invention

[0004] The purpose of the present invention is to provide a substation risk warning method and system based on edge AI analyzer to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A substation risk early warning method based on edge AI analyzer, the method comprising:

[0007] Obtaining equipment distribution information of the substation, and determining the gun position and gun parameters thereof according to the equipment distribution information;

[0008] Determine the position of the edge AI analyzer and its application scope according to the determined gun position; the application scope includes the identification of the gun to which the edge AI analyzer is connected;

[0009] Based on the edge AI analyzer, the video acquired by the gun is received at regular intervals, the target is located in the video, the target frame is determined, the content in the target frame is identified based on a preset convolution feature table, and the risk type is output; the convolution feature table includes a preset risk type and its convolution kernel;

[0010] When the output of identifying the content in the target box based on the preset convolution feature table is empty, the content in the target box is identified based on the AI ​​recognition model in the edge AI analyzer, and the risk type is output;

[0011] Count the risk types of each gun camera, update the application scope of the edge AI analyzer, and the video reception cycle of each gun camera.

[0012] As a further solution of the present invention: the step of obtaining the equipment distribution information of the substation and determining the gun position and its gun parameters according to the equipment distribution information comprises:

[0013] Obtain the equipment registration list of the substation and query the equipment level and location of each device;

[0014] Determine the impact range and the value of each point in the impact range according to the equipment level;

[0015] With the device location as the center, count the impact range and the values ​​of each point in the impact range, and create a monitoring area with numerical values;

[0016] Superimpose the monitoring areas of all devices to obtain the values ​​at each point in the substation area;

[0017] The substation area is divided according to the inscribed rectangle of the monitoring range of the gun camera to obtain sub-areas;

[0018] Accumulating the values ​​at each point in the sub-area to obtain a numerical sum, and determining the gun clarity according to the numerical sum;

[0019] The values ​​of each point in the impact range are:

[0020] Where N is the center value determined by the equipment level, R is the range radius determined by the equipment level, I(ρ) represents the value of the point at the radius ρ; α is the preset correction coefficient.

[0021] As a further solution of the present invention: the step of determining the position of the edge AI analyzer and its application range according to the determined gun position includes:

[0022] Determine the resource proportion of the gun according to the gun clarity;

[0023] Clustering the gun machines according to the resource proportions;

[0024] Calculate the cluster center of each type of gun machine and select the cluster center as the location of the edge AI analyzer;

[0025] Get the label of the gun machine of the corresponding class of the edge AI analyzer as the application scope of the edge AI analyzer.

[0026] As a further solution of the present invention: the step of receiving the video acquired by the gun camera based on the edge AI analyzer at a fixed time, locating the target in the video, determining the target frame, identifying the content in the target frame based on a preset convolution feature table, and outputting the risk type includes:

[0027] Based on the edge AI analyzer, the video captured by the gun camera is received regularly and converted into an image sequence;

[0028] Subtract adjacent images in sequence to obtain a difference image;

[0029] Traversing the pixel points in the difference image, marking the pixel points whose pixel values ​​are greater than a preset threshold, and connecting the adjacent marked pixel points to obtain the target area;

[0030] Eliminate the target area whose area is smaller than the preset area threshold to obtain the target frame;

[0031] The content in the target frame is identified based on the preset convolution feature table to obtain the risk type of each image.

[0032] As a further solution of the present invention: when the output of identifying the content in the target box based on the preset convolution feature table is empty, the step of identifying the content in the target box based on the AI ​​recognition model in the edge AI analyzer and outputting the risk type includes:

[0033] When the output of identifying the content in the target box based on the preset convolution feature table is empty, the image containing the target box is input into the AI ​​recognition model in the edge AI analyzer;

[0034] Read the output of the AI ​​recognition model as the risk type of the target box.

[0035] As a further solution of the present invention: the steps of counting the risk types of each gun camera, updating the application scope of the edge AI analyzer and the video receiving cycle of each gun camera include:

[0036] Read the risk type of the gun based on the image sequence;

[0037] Querying a risk value corresponding to the risk type in a preset value table, and adjusting the video receiving period of the gun camera according to the risk value;

[0038] Determine a correction rate according to the risk value and a preset risk value threshold, adjust the resource ratio of the gun machine based on the correction rate, and update the application scope of the edge AI analyzer;

[0039] The correction rate is determined as follows:

[0040] P=A(F-F0); where P is the correction rate, F is the risk value, F0 is the risk value threshold, and A is a preset constant.

[0041] The technical solution of the present invention also provides a system applied to the above-mentioned substation risk early warning method based on edge AI analyzer, and the system includes:

[0042] A gun installation module is used to obtain the equipment distribution information of the substation and determine the gun position and its gun parameters according to the equipment distribution information;

[0043] An analyzer calibration module, used to determine the position of the edge AI analyzer and its application range according to the determined gun position; the application range includes the identification of the gun to which the edge AI analyzer is connected;

[0044] A target positioning module is used to periodically receive the video acquired by the gun camera based on the edge AI analyzer, locate the target in the video, determine the target frame, identify the content in the target frame based on a preset convolution feature table, and output the risk type; the convolution feature table includes the preset risk type and its convolution kernel;

[0045] A content recognition module, which is used to recognize the content in the target box based on the AI ​​recognition model in the edge AI analyzer and output the risk type when the output of recognizing the content in the target box based on the preset convolution feature table is empty;

[0046] The process update module is used to count the risk types of each gun camera, update the application scope of the edge AI analyzer, and the video receiving cycle of each gun camera.

[0047] As a further solution of the present invention: the gun mounting module includes:

[0048] The equipment query unit is used to obtain the equipment registration list of the substation and query the equipment level and equipment location of each device;

[0049] A value determination unit, used to determine the impact range and the value of each point in the impact range according to the device level;

[0050] The monitoring area determination unit is used to take the device location as the center, count the values ​​of the impact range and each point in the impact range, and create a monitoring area containing values;

[0051] The monitoring area superposition unit is used to superimpose the monitoring areas of all devices to obtain the values ​​at each point in the substation area;

[0052] The area segmentation unit is used to segment the substation area according to the inscribed rectangle of the monitoring range of the gun camera to obtain sub-areas;

[0053] A numerical accumulating unit, used for accumulating the values ​​at each point in the sub-area to obtain a numerical sum, and determining the gun clarity according to the numerical sum;

[0054] The values ​​of each point in the impact range are:

[0055] ; In the formula, N is the center value determined by the equipment level, R is the range radius determined by the equipment level, I(ρ) represents the value of the point at radius ρ; α is the preset correction coefficient.

[0056] As a further solution of the present invention: the analyzer calibration module includes:

[0057] A resource ratio determination unit, used to determine the resource ratio of the gun according to the gun clarity;

[0058] A gun machine clustering unit, used for clustering gun machines according to the resource proportion;

[0059] The position setting unit is used to calculate the cluster center of each type of gun machine and select the cluster center as the position of the edge AI analyzer;

[0060] The label query unit is used to obtain the label of the gun machine of the corresponding class of the edge AI analyzer as the application scope of the edge AI analyzer.

[0061] As a further solution of the present invention: the target positioning module includes:

[0062] An image sequence generation unit, used to convert the video acquired by the gun camera into an image sequence based on the timed reception of the video by the edge AI analyzer;

[0063] An image difference unit, used for sequentially difference adjacent images to obtain a difference image;

[0064] A pixel connection unit, used to traverse the pixel points in the difference image, mark the pixel points whose pixel values ​​are greater than a preset threshold, and connect the adjacent marked pixel points to obtain a target area;

[0065] An area elimination unit is used to eliminate target areas whose areas are smaller than a preset area threshold to obtain a target frame;

[0066] The content recognition unit is used to recognize the content in the target frame based on a preset convolution feature table to obtain the risk type of each image.

[0067] Compared with the prior art, the beneficial effects of the present invention are: the present invention determines the distribution information of the gun machine according to the power equipment, installs the edge AI analyzer according to the gun machine distribution information, and builds a local recognition module based on the convolution kernel table and an AI recognition module based on the AI ​​model into the edge AI analyzer to identify the risks of the substation in a gradient manner, with a faster recognition speed and a wider recognition breadth. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0069] Figure 1 The flowchart of the substation risk warning method based on edge AI analyzer.

[0070] Figure 2 This is a structural block diagram of the substation risk warning system based on edge AI analyzer. DETAILED DESCRIPTION

[0071] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0072] Figure 1 The flowchart of the substation risk warning method based on the edge AI analyzer is as follows. In an embodiment of the present invention, a substation risk warning method based on the edge AI analyzer includes:

[0073] Step S100: Acquire equipment distribution information of the substation, and determine the gun position and gun parameters according to the equipment distribution information;

[0074] A substation refers to a place in the power system where voltage and current are transformed, electric energy is received and distributed. It is equipped with a variety of power equipment. The location and parameters of each power equipment are the equipment distribution information. By obtaining the distribution information of the power equipment, the installation location and installation parameters of the gun camera can be determined. The gun camera can be understood as a fixed camera with high precision and a fixed shooting frame. Correspondingly, the gun camera parameters are the clarity of the camera.

[0075] Step S200: determining the position of the edge AI analyzer and its application scope according to the determined gun position; the application scope includes the identification of the gun to which the edge AI analyzer is connected;

[0076] There is a data interaction channel between the edge AI analyzer and the gun. The position of the edge AI analyzer can be determined according to the determined gun position. The gun corresponding to each edge AI analyzer is the application scope of the edge AI analyzer; the application scope can be composed of the identification of the gun connected to the edge AI analyzer, and the identification is generally a number.

[0077] Step S300: Based on the edge AI analyzer, the video acquired by the gun camera is received at regular intervals, the target is located in the video, the target frame is determined, the content in the target frame is identified based on a preset convolution feature table, and the risk type is output; the convolution feature table includes a preset risk type and its convolution kernel;

[0078] For each edge AI analyzer, there is data interaction between it and a part of the gun cameras. It receives the videos acquired by the gun cameras at regular intervals, locates the target in each video, obtains the target frame, and first identifies the content in the target frame based on the preset convolution feature table to obtain the risk type. This process is essentially an image traversal and comparison process and does not involve the AI ​​recognition process. The convolution kernel table is a preset table. The staff pre-acquires images of different risk types, identifies the images, and then extracts image features, which are called convolution kernels.

[0079] Step S400: When the output of identifying the content in the target box based on the preset convolution feature table is empty, identifying the content in the target box based on the AI ​​recognition model in the edge AI analyzer and outputting the risk type;

[0080] The target recognition process based on the convolution feature table can be performed locally. In the existing technology, the recognition speed may not be much different, but the amount of resources consumed is less. The disadvantage of the target recognition process based on the convolution feature table is that it is impossible to identify risk types that have not been counted in advance. At this time, after introducing the AI ​​recognition model, unknown targets can be identified, which greatly improves the comprehensiveness of recognition.

[0081] Step S500: Count the risk types of each gun camera, update the application scope of the edge AI analyzer and the video receiving cycle of each gun camera;

[0082] After identifying the risk type, the substation risk warning process has been completed. On this basis, considering the working pressure of each edge AI analyzer, this application introduces an update process. For a certain edge AI analyzer, the risk types corresponding to each gun camera are counted. The more risk types there are, the shorter the video receiving cycle of the corresponding gun camera will be, and the more videos will be obtained. Correspondingly, the number of gun cameras connected to the edge AI analyzer will decrease.

[0083] As a preferred embodiment of the technical solution of the present invention, the step of obtaining the equipment distribution information of the substation and determining the gun position and its gun parameters according to the equipment distribution information includes:

[0084] Obtain the equipment registration list of the substation and query the equipment level and location of each device;

[0085] Determine the impact range and the value of each point in the impact range according to the equipment level;

[0086] With the device location as the center, count the impact range and the values ​​of each point in the impact range, and create a monitoring area with numerical values;

[0087] Superimpose the monitoring areas of all devices to obtain the values ​​at each point in the substation area;

[0088] The substation area is divided according to the inscribed rectangle of the monitoring range of the gun camera to obtain sub-areas;

[0089] Accumulating the values ​​at each point in the sub-area to obtain a numerical sum, and determining the gun clarity according to the numerical sum;

[0090] The values ​​of each point in the impact range are:

[0091] ; In the formula, N is the center value determined by the equipment level, R is the range radius determined by the equipment level, I(ρ) represents the value of the point at radius ρ; α is the preset correction coefficient.

[0092] The value of each point can also be applied with a two-dimensional Gaussian distribution model, which has higher smoothness.

[0093] In an example of the technical solution of the present invention, the power equipment in the substation is stored in an equipment registration table, including the equipment location and equipment level of each power equipment. The equipment level represents the importance of the power equipment. The impact range is determined based on the equipment level. The higher the equipment level, the larger the impact range. Combined with the equipment location, an impact area can be determined, which is called a monitoring area. The monitoring area is generally circular.

[0094] By superimposing the monitoring areas of all devices, we can obtain the values ​​of each point in the substation area. The value at each point reflects the importance of the corresponding point. Based on this, we can obtain the monitoring range of the gun camera. This application requires global monitoring of the entire substation. The set of monitoring ranges of all gun cameras must be larger than the substation area. Therefore, the entire area of ​​the substation is divided by the monitoring range of the gun camera to obtain sub-areas.

[0095] Since a numerical value representing the importance has been calculated for each point, the values ​​at each point in the sub-area can be accumulated to obtain a numerical sum, and the gun clarity can be determined based on the numerical sum; the gun clarity is proportional to the numerical sum; that is, the more important the area corresponding to the gun is, the higher the clarity will be.

[0096] As a preferred embodiment of the technical solution of the present invention, the step of determining the position of the edge AI analyzer and its application range according to the determined gun position includes:

[0097] Determine the resource proportion of the gun according to the gun clarity;

[0098] Clustering the gun machines according to the resource proportions;

[0099] Calculate the cluster center of each type of gun machine and select the cluster center as the location of the edge AI analyzer;

[0100] Get the label of the gun machine of the corresponding class of the edge AI analyzer as the application scope of the edge AI analyzer.

[0101] The higher the definition of the gun camera, the higher the amount of resources required to process the corresponding video, and the higher the resource share. By clustering the gun cameras according to the resource share, multiple categories of gun cameras can be obtained. For a category of gun cameras corresponding to a higher resource share, the performance of the selected edge AI analyzer will be higher. On the contrary, for a category of gun cameras corresponding to a lower resource share, the performance of the selected edge AI analyzer will be lower. In order to minimize the transmission pressure of the data transmission process (between the gun camera and the edge AI analyzer), the cluster center is selected as the position of the edge AI analyzer, and the label of the gun camera of the corresponding category of the edge AI analyzer is obtained as the application scope of the edge AI analyzer.

[0102] As a preferred embodiment of the technical solution of the present invention, the step of receiving the video acquired by the gun camera based on the edge AI analyzer at regular intervals, locating the target in the video, determining the target frame, identifying the content in the target frame based on a preset convolution feature table, and outputting the risk type includes:

[0103] Based on the edge AI analyzer, the video captured by the gun camera is received regularly and converted into an image sequence;

[0104] Subtract adjacent images in sequence to obtain a difference image;

[0105] Traversing the pixel points in the difference image, marking the pixel points whose pixel values ​​are greater than a preset threshold, and connecting the adjacent marked pixel points to obtain the target area;

[0106] Eliminate the target area whose area is smaller than the preset area threshold to obtain the target frame;

[0107] The content in the target frame is identified based on the preset convolution feature table to obtain the risk type of each image.

[0108] In an example of the technical solution of the present invention, a specific description is given of the target positioning process and the application process of the convolution feature table. Based on the edge AI analyzer, the video acquired by the gun is received periodically, the video is converted into an image sequence (the audio is removed), and adjacent images are subtracted in turn to obtain a difference image. For the monitoring video, only the pixels corresponding to the moving object will have differences, and the other pixels are close to zero after subtraction; the pixels in the difference image are traversed, the pixels with pixel values ​​greater than a preset threshold are marked, and the adjacent marked pixels are connected to obtain the target area. The target area is the active area where the moving object is located. The target area (noise) with a particularly small area is removed, and the remaining target area is used as the area of ​​the real moving object. The matrix shape of the area is obtained, which is called the target frame; the content in the target frame is identified based on the preset convolution feature table. After the match is successful, the risk type corresponding to the target frame is obtained.

[0109] As a preferred embodiment of the technical solution of the present invention, when the output of identifying the content in the target box based on the preset convolution feature table is empty, the step of identifying the content in the target box based on the AI ​​recognition model in the edge AI analyzer and outputting the risk type includes:

[0110] When the output of identifying the content in the target box based on the preset convolution feature table is empty, the image containing the target box is input into the AI ​​recognition model in the edge AI analyzer;

[0111] Read the output of the AI ​​recognition model as the risk type of the target box.

[0112] When the output of identifying the content in the target box based on the preset convolution feature table is empty, that is, when identifying the content in the target box based on the preset convolution feature table, the match fails. At this time, the AI ​​recognition model in the edge AI analyzer is applied to identify the target. On the basis that the AI ​​capability is already strong enough, the recognition breadth and recognition depth of the present application are greatly improved.

[0113] It is worth mentioning that in actual applications, since there are only a limited number of risk types, the convolutional feature table is sufficient to cover most types, and the application frequency of AI recognition models is actually very low.

[0114] As a preferred embodiment of the technical solution of the present invention, the steps of counting the risk types of each gun camera, updating the application scope of the edge AI analyzer and the video receiving cycle of each gun camera include:

[0115] Read the risk type of the gun based on the image sequence;

[0116] Querying a risk value corresponding to the risk type in a preset value table, and adjusting the video receiving period of the gun camera according to the risk value;

[0117] Determine a correction rate according to the risk value and a preset risk value threshold, adjust the resource ratio of the gun machine based on the correction rate, and update the application scope of the edge AI analyzer;

[0118] The correction rate is determined as follows:

[0119] P=A(F-F0); where P is the correction rate, F is the risk value, F0 is the risk value threshold, and A is a preset constant.

[0120] In an example of the technical solution of the present invention, a negative feedback adjustment scheme is provided, which reads the risk type of the gun camera based on the image sequence, and the staff presets a numerical value representing the degree of risk for each risk type, that is, the risk value, and queries the risk value corresponding to the risk type. The video receiving period of the gun camera is adjusted according to the risk value, and the video receiving period is inversely proportional to the risk value. The larger the risk value, the shorter the video receiving period; further, the correction rate is determined according to the risk value, and the resource ratio of the gun camera is adjusted, which affects the clustering process of the gun camera, that is, which edge AI analyzers manage different gun cameras. The performance of different edge AI analyzers is different. The higher the risk value, the higher the corresponding resource ratio.

[0121] Figure 2 : is a structural block diagram of a substation risk warning system based on an edge AI analyzer. In an embodiment of the present invention, a system applied to the above-mentioned substation risk warning method based on an edge AI analyzer, the system 10 includes:

[0122] The gun installation module 11 is used to obtain the equipment distribution information of the substation and determine the gun position and its gun parameters according to the equipment distribution information;

[0123] An analyzer calibration module 12, used to determine the position of the edge AI analyzer and its application range according to the determined gun position; the application range includes the identification of the gun to which the edge AI analyzer is connected;

[0124] The target positioning module 13 is used to periodically receive the video acquired by the gun based on the edge AI analyzer, locate the target in the video, determine the target frame, identify the content in the target frame based on a preset convolution feature table, and output the risk type; the convolution feature table includes the preset risk type and its convolution kernel;

[0125] A content recognition module 14 is used to recognize the content in the target box based on the AI ​​recognition model in the edge AI analyzer and output the risk type when the output of recognizing the content in the target box based on the preset convolution feature table is empty;

[0126] The process update module 15 is used to count the risk types of each gun camera, update the application scope of the edge AI analyzer and the video receiving cycle of each gun camera.

[0127] Furthermore, the gun mounting module 11 includes:

[0128] The equipment query unit is used to obtain the equipment record list of the substation and query the equipment level and equipment location of each equipment;

[0129] A value determination unit, used to determine the impact range and the value of each point in the impact range according to the device level;

[0130] The monitoring area determination unit is used to take the device location as the center, count the values ​​of the impact range and each point in the impact range, and create a monitoring area containing values;

[0131] The monitoring area superposition unit is used to superimpose the monitoring areas of all devices to obtain the values ​​at each point in the substation area;

[0132] The area segmentation unit is used to segment the substation area according to the inscribed rectangle of the monitoring range of the gun camera to obtain sub-areas;

[0133] A numerical accumulating unit, used for accumulating the values ​​at each point in the sub-area to obtain a numerical sum, and determining the gun clarity according to the numerical sum;

[0134] The values ​​of each point in the impact range are:

[0135] ; In the formula, N is the center value determined by the equipment level, R is the range radius determined by the equipment level, I(ρ) represents the value of the point at radius ρ; α is the preset correction coefficient.

[0136] Specifically, the analyzer calibration module 12 includes:

[0137] A resource ratio determination unit, used to determine the resource ratio of the gun according to the gun clarity;

[0138] A gun machine clustering unit, used for clustering gun machines according to the resource proportion;

[0139] The position setting unit is used to calculate the cluster center of each type of gun machine and select the cluster center as the position of the edge AI analyzer;

[0140] The label query unit is used to obtain the label of the gun machine of the corresponding class of the edge AI analyzer as the application scope of the edge AI analyzer.

[0141] In addition, the target positioning module 13 includes:

[0142] An image sequence generation unit, used to convert the video acquired by the gun camera into an image sequence based on the timed reception of the video by the edge AI analyzer;

[0143] An image difference unit, used for sequentially difference adjacent images to obtain a difference image;

[0144] A pixel connection unit, used to traverse the pixel points in the difference image, mark the pixel points whose pixel values ​​are greater than a preset threshold, and connect the adjacent marked pixel points to obtain a target area;

[0145] An area elimination unit is used to eliminate target areas whose areas are smaller than a preset area threshold to obtain a target frame;

[0146] The content recognition unit is used to recognize the content in the target frame based on a preset convolution feature table to obtain the risk type of each image.

[0147] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A substation risk warning method based on edge AI analyzer, characterized in that: The method comprises: Step S100: Acquire equipment distribution information of the substation, and determine the gun position and gun parameters according to the equipment distribution information; Step S200: determining the position of the edge AI analyzer and its application scope according to the determined gun position; the application scope includes the identification of the gun to which the edge AI analyzer is connected; Step S300: Based on the edge AI analyzer, the video acquired by the gun camera is received at regular intervals, the target is located in the video, the target frame is determined, the content in the target frame is identified based on a preset convolution feature table, and the risk type is output; the convolution feature table includes a preset risk type and its convolution kernel; Step S400: When the output of identifying the content in the target box based on the preset convolution feature table is empty, identifying the content in the target box based on the AI ​​recognition model in the edge AI analyzer and outputting the risk type; Step S500: Count the risk types of each gun camera, update the application scope of the edge AI analyzer and the video receiving cycle of each gun camera.

2. The substation risk early warning method based on edge AI analyzer according to claim 1 is characterized in that: The step of obtaining the equipment distribution information of the substation and determining the gun position and its gun parameters according to the equipment distribution information comprises: Obtain the equipment registration list of the substation and query the equipment level and location of each device; Determine the impact range and the value of each point in the impact range according to the equipment level; With the device location as the center, count the impact range and the values ​​of each point in the impact range, and create a monitoring area with numerical values; Superimpose the monitoring areas of all devices to obtain the values ​​at each point in the substation area; The substation area is divided according to the inscribed rectangle of the monitoring range of the gun camera to obtain sub-areas; Accumulating the values ​​at each point in the sub-area to obtain a numerical sum, and determining the gun clarity according to the numerical sum; The values ​​of each point in the impact range are: Where N is the center value determined by the equipment level, R is the range radius determined by the equipment level, I(ρ) represents the value of the point at the radius ρ; α is the preset correction coefficient.

3. The substation risk early warning method based on edge AI analyzer according to claim 1 is characterized in that: The step of determining the position of the edge AI analyzer and its application range according to the determined gun position includes: Determine the resource proportion of the gun according to the gun clarity; Clustering the gun machines according to the resource proportions; Calculate the cluster center of each type of gun machine and select the cluster center as the location of the edge AI analyzer; Get the label of the gun machine of the corresponding class of the edge AI analyzer as the application scope of the edge AI analyzer.

4. The substation risk early warning method based on edge AI analyzer according to claim 1 is characterized in that: The step of receiving the video acquired by the gun camera at a fixed time based on the edge AI analyzer, locating the target in the video, determining the target frame, identifying the content in the target frame based on a preset convolution feature table, and outputting the risk type includes: Based on the edge AI analyzer, the video captured by the gun camera is received regularly and converted into an image sequence; Subtract adjacent images in sequence to obtain a difference image; Traversing the pixel points in the difference image, marking the pixel points whose pixel values ​​are greater than a preset threshold, and connecting the adjacent marked pixel points to obtain the target area; Eliminate the target area whose area is smaller than the preset area threshold to obtain the target frame; The content in the target frame is identified based on the preset convolution feature table to obtain the risk type of each image.

5. The substation risk early warning method based on edge AI analyzer according to claim 1 is characterized in that: When the output of identifying the content in the target box based on the preset convolution feature table is empty, the step of identifying the content in the target box based on the AI ​​recognition model in the edge AI analyzer and outputting the risk type includes: When the output of identifying the content in the target box based on the preset convolution feature table is empty, the image containing the target box is input into the AI ​​recognition model in the edge AI analyzer; Read the output of the AI ​​recognition model as the risk type of the target box.

6. The substation risk early warning method based on edge AI analyzer according to claim 1 is characterized in that: The steps of counting the risk types of each gun camera, updating the application scope of the edge AI analyzer and the video receiving cycle of each gun camera include: Read the risk type of the gun based on the image sequence; Querying a risk value corresponding to the risk type in a preset value table, and adjusting the video receiving period of the gun camera according to the risk value; Determine a correction rate according to the risk value and a preset risk value threshold, adjust the resource ratio of the gun machine based on the correction rate, and update the application scope of the edge AI analyzer; The correction rate is determined as follows: P=A(F-F0); where P is the correction rate, F is the risk value, F0 is the risk value threshold, and A is a preset constant.

7. Substation risk warning system based on edge AI analyzer, characterized by: The substation risk warning method based on the edge AI analyzer is applied to any one of claims 1 to 6, and the substation risk warning system (10) based on the edge AI analyzer comprises: A gun installation module (11) is used to obtain equipment distribution information of the substation and determine the gun position and gun parameters according to the equipment distribution information; An analyzer calibration module (12), used to determine the position of the edge AI analyzer and its application range according to the determined gun position; the application range includes the identification of the gun to which the edge AI analyzer is connected; A target positioning module (13) is used to periodically receive the video acquired by the gun camera based on the edge AI analyzer, perform target positioning on the video, determine the target frame, identify the content in the target frame based on a preset convolution feature table, and output the risk type; the convolution feature table includes a preset risk type and its convolution kernel; A content recognition module (14), configured to recognize the content in the target frame based on an AI recognition model in an edge AI analyzer and output a risk type when the output of recognizing the content in the target frame based on a preset convolution feature table is empty; The process update module (15) is used to count the risk types of each gun camera, update the application scope of the edge AI analyzer and the video receiving cycle of each gun camera.

8. The substation risk early warning system based on edge AI analyzer according to claim 7 is characterized in that: The gun mounting module (11) comprises: The equipment query unit is used to obtain the equipment record list of the substation and query the equipment level and equipment location of each equipment; A value determination unit, used to determine the impact range and the value of each point in the impact range according to the device level; The monitoring area determination unit is used to take the device location as the center, count the values ​​of the impact range and each point in the impact range, and create a monitoring area containing values; The monitoring area superposition unit is used to superimpose the monitoring areas of all devices to obtain the values ​​at each point in the substation area; The area segmentation unit is used to segment the substation area according to the inscribed rectangle of the monitoring range of the gun camera to obtain sub-areas; A numerical accumulating unit, used for accumulating the values ​​at each point in the sub-area to obtain a numerical sum, and determining the gun clarity according to the numerical sum; The values ​​of each point in the impact range are: Where N is the center value determined by the equipment level, R is the range radius determined by the equipment level, I(ρ) represents the value of the point at the radius ρ; α is the preset correction coefficient.

9. The substation risk early warning system based on edge AI analyzer according to claim 7 is characterized in that: The analyzer calibration module (12) comprises: A resource ratio determination unit, used to determine the resource ratio of the gun according to the gun clarity; A gun machine clustering unit, used for clustering gun machines according to the resource proportion; The position setting unit is used to calculate the cluster center of each type of gun machine and select the cluster center as the position of the edge AI analyzer; The label query unit is used to obtain the label of the gun machine of the corresponding class of the edge AI analyzer as the application scope of the edge AI analyzer.

10. The substation risk early warning system based on edge AI analyzer according to claim 7 is characterized in that: The target positioning module (13) comprises: An image sequence generation unit, used to convert the video acquired by the gun camera into an image sequence based on the timed reception of the video by the edge AI analyzer; An image difference unit, used for sequentially difference adjacent images to obtain a difference image; A pixel connection unit, used to traverse the pixel points in the difference image, mark the pixel points whose pixel values ​​are greater than a preset threshold, and connect the adjacent marked pixel points to obtain a target area; An area elimination unit is used to eliminate target areas whose areas are smaller than a preset area threshold to obtain a target frame; The content recognition unit is used to recognize the content in the target frame based on a preset convolution feature table to obtain the risk type of each image.

Images